Think im gonna go for ipisoft, im liking the fact you can use 1 computer with x2 kinect 2s. It has export skeleton for iclone. The only drawback is price.

Just wondering has anyone used both. Any major drawbacks i should know about?

Thanks in advance  

Mike Kelley has tried out ipisoft years ago and his conclusion was that nothing beats a mocap suit such as Perception Neuron.

---

Ah man. Im down to selling scraps of my defunked printing business on ebay and no one is biting. 

Think after forking out over 2k in 12 months on reallusion im a bit stuck. (I only had ic7+cc3 for 11 months when the upgrade kicked in).

I got x2 kinects x 2 adapters x 2 usb cards and a subscription for around 600.

Them suits are about 2k. 

Ive been watching loads of vids on this mocap. Im sure a hell of a lot of work is in post. Seen a funny vid where sum vlogger on youtube is ranting cos he spent 4k and his results are really poor.  

Yeah, I was about spending $$$ for suit but with the way to fix footsliding (major issue for AI mocap like threeD tracker and ipisoft ) coming with iclone 8 I'll definitely keep my money. If you can, buy it but today, for me, possibilities are more than last year 


     

I understand money concerns.

I've played a bit with the rotoscoping option in iClone (animating from video) and the motion correction such as for foot-sliding as Charly mentions, is indeed pretty good.

And since you already have some of the equipment you could try the demo and see how it goes. 

---

Honestly I would not bother with either ipisoft or brekel (and I had ipisoft for years - it's just WAY too much post work).

If money is an issue to the point where you can't afford a mocap suit, then I would look into those services where you upload your video (just a video of yourself) and it turns it into mocap.  There is a cost there as well, and the better ones won't allow you to do too much without spending $$$, but for a small cost you can do SOME mocap and it will be far better until you can actually afford your own mocap suit.

Then again, some folks just have to learn the hard way - the thing about mocap is that it's SO seductive in looking at samples folks have slaved over.  In reality, until you get a mocap suit, you need a TON of time and patience.  And in that case, if you have those, nothing beats hand animation.

It's like the age old story of us programmers.  We used to come to clients and write "Good, fast, cheap" on the blackboard and tell them - "Pick any two".  The same is true with motion capture.  You can get it good and fast, but it won't be cheap.  Or you can get it cheap and good, but it will take FOREVER.  Or you can get it fast and cheap, but it will be shit.  Again - pick any two.

---

Alienware Aurora R16, Win 11, i9-149000KF, 3.20GHz CPU, 64GB RAM, RTX 4090 (24GB), Samsung 870 Pro 8TB, Gen3 MVNe M-2 SSD, 4TBx2, 39" Alienware Widescreen Monitor
Mike "ex-genius" Kelley

Wait - wasn't there a guy here who was selling his PN suit?

You might email him and ask.  The cost might be in your budget (and you can always haggle with him).

You will thank me later (or you will just learn the hard way.  As I said, some folks just need to go through it for themselves.  Nothing wrong with that if you are young, as time is the thing you have the most).

---

Alienware Aurora R16, Win 11, i9-149000KF, 3.20GHz CPU, 64GB RAM, RTX 4090 (24GB), Samsung 870 Pro 8TB, Gen3 MVNe M-2 SSD, 4TBx2, 39" Alienware Widescreen Monitor
Mike "ex-genius" Kelley

Had a play with rotoscope. Is a bit frustrating. Reallusion could make it a goldmine by introducing Opacity, especially on the wireframe. The 3d model over the video always gets in the way. 

I was wondering if they could remove (visually) the main joints of the 3d character. You could easily align the joints in the video behind.
When filming your video you could even set up color markers on your elbow, shoulder, wrists, knees and ankles and then align them up with the opacity removed from the model/wireframe. 

What I've been doing is a hybrid: Use one of the AI online services that create a mocap from a video and then use the rotoscoping feature for refinement and cleaning up. It is work of course, but it's doable and save a lot of money.

---

Do u not think the 3d model is just in the way of the video?. I find it quite frustrating. I found a hack by freedom which lets you turn down the opacity of 3d model. But this creates a bit of a mess because it makes a ghost like opacity on clothes, hair, skin....
The wireframe feature is useful , if u could drop the opacity on that and change the color i think it could be much easier to use.

Think i may place it in the 'ideas' section.

I just cancelled deepmotion because the end result was so shakey and nothing like my vids. I then had a go at rotoscope and realised just how bad i am at animation (looks like old stop animation)... so sought out a new solution.